U.S. patent application number 13/877059 was filed with the patent office on 2013-08-15 for automated health data acquisition, processing and communication system.
This patent application is currently assigned to dacadoo ag. The applicant listed for this patent is Laurence Jacobs, David Leason, Andre Naef, Peter Ohnemus. Invention is credited to Laurence Jacobs, David Leason, Andre Naef, Peter Ohnemus.
Application Number | 20130211858 13/877059 |
Document ID | / |
Family ID | 45938652 |
Filed Date | 2013-08-15 |
United States Patent
Application |
20130211858 |
Kind Code |
A1 |
Ohnemus; Peter ; et
al. |
August 15, 2013 |
AUTOMATED HEALTH DATA ACQUISITION, PROCESSING AND COMMUNICATION
SYSTEM
Abstract
A unique health score computation method is disclosed which
masks underlying health statistics, yet provides a benchmark for a
variety of applications. A system and method for collecting health
related information, processing the information into a composite
numerical value, and publishing the value is provided. The system
includes a computer having a processor, memory, and code modules
executing in the processor for implementation of the method.
Information concerning a plurality of intrinsic and extrinsic
parameters of a user is collected. Weighting factors are applied to
the parameter in order control the relative affect each parameter
has on the user's calculated numerical. The health score is
computed using the processor by combining the weighted parameters
in accordance with an algorithm. The numerical value is published
to a designated group via a portal, while the underlying parameters
remain private. In one implementation, the portal is an internet
based information sharing forum.
Inventors: |
Ohnemus; Peter; (Kusnacht,
CH) ; Naef; Andre; (Zurich, CH) ; Jacobs;
Laurence; (Thalwil, CH) ; Leason; David;
(Chappaqua, NY) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Ohnemus; Peter
Naef; Andre
Jacobs; Laurence
Leason; David |
Kusnacht
Zurich
Thalwil
Chappaqua |
NY |
CH
CH
CH
US |
|
|
Assignee: |
dacadoo ag
Zurich
CH
|
Family ID: |
45938652 |
Appl. No.: |
13/877059 |
Filed: |
September 29, 2011 |
PCT Filed: |
September 29, 2011 |
PCT NO: |
PCT/US11/53971 |
371 Date: |
April 23, 2013 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61387906 |
Sep 29, 2010 |
|
|
|
61495247 |
Jun 9, 2011 |
|
|
|
Current U.S.
Class: |
705/3 |
Current CPC
Class: |
G16H 10/60 20180101;
G16H 50/30 20180101; G16H 20/30 20180101 |
Class at
Publication: |
705/3 |
International
Class: |
G06F 19/00 20060101
G06F019/00 |
Claims
1. A computer implemented method for processing private health
related data into a masked numerical score suitable for publishing,
comprising the steps of: receiving data into a memory on a
plurality of intrinsic medical parameters and extrinsic physical
activity parameters of a user; storing the received data in the
memory; storing weighting factors in the memory; processing the
received data by executing code in a processor that configures the
processor to apply the weighting factors to the intrinsic medical
parameters and the extrinsic physical activity parameters, wherein
the weighting factors for the extrinsic physical activity
parameters include a decay component arranged to reduce the
relative weight of the extrinsic physical activity parameters for a
physical activity in dependence on at least one factor associated
with the user; transforming the processed data concerning the
intrinsic medical parameters and the extrinsic physical activity
parameters by executing code in the processor into a masked
composite numerical value in which the code is operative to combine
the weighted parameters in accordance with an algorithm; and
automatically publishing the masked composite numerical value to a
designated group via a portal, using code executing in the
processor and free of human intervention, while maintaining the
collected information concerning the intrinsic medical parameters
and the extrinsic physical activity parameters private.
2. The method of claim 1, wherein the factor associated with the
user is an age or an age range of the user such that the decay
component reduces the relative weight of the extrinsic physical
activity parameters for a first user of a first age or age range
differently than a second user of a second age or age range.
3. The method of claim 1, further comprising the step of averaging
the published masked composite numerical value of a group of users
to determine a group composite numerical value using further code
executing in the processor.
4. The method of claim 1, further comprising the steps of:
receiving data into the memory on extrinsic lifestyle parameters of
a user, wherein the step of processing includes executing
additional code in the processor that configures the processor to
apply weighting factors to the extrinsic lifestyle parameters and
includes a decay component further arranged to reduce the relative
weight of the extrinsic lifestyle parameters in dependence on at
least one factor associated with the user, wherein the step of
transforming the processed data includes using the processor by
combining the weighted intrinsic medical parameters, extrinsic
physical activity parameters and extrinsic lifestyle parameters
with the algorithm.
5. The method of claim 1, wherein the steps of processing,
transforming and publishing are performed substantially
automatically upon receipt of data on intrinsic medical parameters
or extrinsic parameters of a user.
6. The method of claim 5, additional steps of: communicating either
the processed data or the masked composite numerical value to an
exercise machine and automatically establishing the exercise
program on that basis, and communicating activity information from
the exercise machine to the memory for inclusion among the
extrinsic physical activity parameters.
7. The method of claim 1, further comprising monitoring the
composite numerical value and causing triggering of a feedback
communication by executing code in the processor and without human
intervention.
8. The method of claim 7, wherein the feedback communication is
operative to provide an alert to the user to initiate a physical
activity or change a scheduled physical activity.
9. The method of claim 7, wherein the feedback communication
comprises an alert sent to a predetermined person.
10. The method of claim 7, wherein the step of monitoring comprises
monitoring value over time and triggering alert in dependence on
change over time.
11. The method of claim 7, wherein the step of triggering a
feedback communication comprises sending an electronic
communication directed to the user including directions on changes
to the user's physical activity and/or lifestyle for improving the
masked composite numerical value.
12. The method of claim 7, further comprising calculating, by
executing additional code in the processor, a predicative masked
composite numerical value, which is indicative of a predicted
future state based on past data, using the received data on the
plurality of intrinsic medical parameters and extrinsic physical
activity parameters of the user in accordance with a predicative
algorithm and causing triggering of a predictive feedback
communication.
13. The method of claim 1, wherein the step of processing the
received extrinsic physical activity parameters includes: obtaining
a measure of calories expended in the physical activity into the
memory; and executing further code in the processor that configures
the processor to: transform the measured calories into a metabolic
equivalent, MET, value by dividing by the user's body weight;
divide the MET value between a health pool and a bonus pool,
wherein the bonus pool has a predetermined size and any divided MET
value exceeding the bonus pool size is allocated to the health
pool; and apply a daily decay component to the bonus pool; wherein
the step of transforming the processed data comprises combining the
weighted intrinsic medical parameters and a weighted health pool
value in accordance with the algorithm.
14. A health monitoring system comprising: a communication unit
operable to receive data on a plurality of intrinsic medical
parameters and extrinsic physical activity parameters of a user; a
memory arranged to store the received data and to store weighting
factors; a processor arranged to process the received data by
executing code that configures the processor to apply the weighting
factors to the intrinsic medical parameters and the extrinsic
physical activity parameters, wherein the weighting factors for the
extrinsic physical activity parameters include a decay component
arranged to reduce the relative weight of the physical activity
parameters for a physical activity in dependence on at least one
factor associated with the user; the processor being further
arranged to execute code to transform the processed data concerning
the intrinsic medical parameters and the extrinsic physical
activity parameters into a masked composite numerical value using
the processor by combining the weighted parameters in accordance
with an algorithm; and a portal arranged to publish the masked
composite numerical value to a designated group while maintaining
the collected information concerning the intrinsic medical
parameters and the extrinsic physical activity parameters
private.
15. The system method of claim 14, wherein the factor associated
with the user is an age or an age range of the user such that the
decay component reduces the relative weight of the extrinsic
physical activity parameters for a first user of a first age or age
range differently than a second user of a second age or age
range.
16. The system of claim 14, wherein the communication unit is
further arranged to receiving data on extrinsic lifestyle
parameters of a user, wherein the processor is arranged to execute
code to: apply weighting factors to the extrinsic lifestyle
parameters, apply a decay component that is arranged to reduce the
relative weight of the extrinsic lifestyle parameters in dependence
on at least one factor associated with the user, and transform the
processed data by combining the weighted intrinsic medical
parameters, extrinsic physical activity parameters and extrinsic
lifestyle parameters in dependence on the algorithm.
17. The system of claim 14, arranged to automatically perform said
processing upon receipt of data on intrinsic medical parameters or
extrinsic parameters of a user.
18. The system of claim 17, further comprising a remote user
device, the system being arranged to communicate with the remote
user device during physical activity to receive at least selected
ones of the extrinsic physical activity parameters.
19. The system of claim 14, further comprising a monitoring unit
arranged to monitor the composite numerical values and being
arranged to cause triggering of a feedback communication upon
detecting a predetermined event associated with the monitored
composite numerical values.
20. The system of claim 19, wherein the feedback communication is
operative to re-configure a program to define or that is defining a
scheduled physical activity for the user.
21. The system of claim 19, wherein the monitoring unit is arranged
to cause transmission of an electronic communication directed to
the user including directions on changes to the user's physical
activity and/or lifestyle for improving the masked composite
numerical value.
22. The system of claim 19, wherein the processor is further
arranged to execute code that configures the processor to calculate
a predicative masked composite numerical value that is indicative
of a predicted future state based on past data, using the received
data on the plurality of intrinsic medical parameters and extrinsic
physical activity parameters of the user in accordance with a
predicative algorithm and wherein the monitor is arranged to cause
triggering of a predictive feedback communication.
23. The system of claim 14, wherein the processor is arranged to
process the received extrinsic physical activity parameters by
executing code that configures the processor to perform steps
including: obtaining a measure of calories expended in the physical
activity; transforming the measured calories into a metabolic
equivalent, MET, value by dividing by the user's body weight;
dividing the MET value between a health pool and a bonus pool,
wherein the bonus pool has a predetermined size and any divided MET
value exceeding the bonus pool size is allocated to the health
pool; applying a daily decay component to the bonus pool; and
transforming the processed data comprises combining the weighted
intrinsic medical parameters and a weighted health pool value in
accordance with the algorithm.
24. The system of claim 14, further comprising a bi-directional
communication link to an exercise machine that is configured to:
communicate either the processed data or the masked composite
numerical value to the exercise machine; automatically establishing
the exercise program on the basis of the communicated data or the
masked composite numerical value, and receive from the exercise
machine into the memory activity information for inclusion among
the extrinsic physical activity parameters.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims the benefit of U.S. Patent
Application Ser. No. 61/387,906, filed Sep. 29, 2010, and U.S.
Patent Application Ser. No. 61/495,247, filed Jun. 9, 2011, which
are hereby incorporated by reference in their entireties.
FIELD OF THE INVENTION
[0002] The present invention concerns a computer implemented system
for the acquisition of medical data and its processing for
diagnostic, benchmarking, analytics and redistribution purposes.
More particularly, the invention concerns a computer implemented
system and method for acquisition, diagnosis, benchmarking,
analytics and/or redistribution of medical data.
BACKGROUND OF THE INVENTION
[0003] Despite advances in many areas of technology, there are
still barriers to assessing the relative health of a person in a
rapid, cost effective, and timely manner. With the increase in
health care costs and prevalence of diseases related to unhealthy
lifestyles such as diabetes and heart disease, it is important to
assess the relative health of individuals, and this has not been
adequately addressed. In many areas of the world, access to doctors
is limited. Even in the developed world, a doctor's time is
considered a precious commodity and there are often long waiting
lists and doctor-to-specialist referral systems have to be
navigated before being seen. In more developed countries the ratio
of doctors to the population may be on the order of 1:1,000
persons, while in less developed countries the ratio may be
1:100,000. There are also cost barriers to having access to a
doctor because an appointment with a doctor can be very expensive,
especially if an individual does not have any health insurance or
lacks sufficient coverage. Accordingly, it can be very difficult to
gain access to medical professionals in order to receive
information about one's health.
[0004] Even if an individual had access to his or her health
information, the mechanisms for conveying that information to
others is lacking or non-existent. Privacy laws restrict the type
of information that can be shared and the manner in which it can be
shared. Privacy laws relating to health information are
particularly strict in regard to the information that can be
shared. This is to protect a person from disclosure of sensitive
information. Accordingly, the sharing of health related information
is generally discouraged. It is also difficult to share health
related information with friends and family. Often health
information is only verbally conveyed by a doctor to a patient, or
the patient will only receive paper copies of lab test results.
Systems are lacking for easily sharing such information with
others, especially with large groups of persons located in
geographically remote locations.
[0005] Prior art systems that provide a limited type of numerical
score which is related to a person's health have been disclosed.
For example, U.S. Patent Publication No. 2009/0105550 to Rothman et
al. discloses a system and method for providing a health score for
a patient. However, this disclosure is primarily directed to
calculating a health score of a patient in a hospital, post
surgery, and the health score is based on medical data measured
from the patient (e.g., blood pressure, temperature, respiration,
etc.). This method fails to take into account the extrinsic
activities of the patient, such as the daily physical exercise
activities of the patient. U.S. Patent Publication No. 2005/0228692
to Hodgdon discloses a system that calculates a health score based
on measured medical data and can include a self assessment survey,
which can include surveying a participant's exercise habits.
However, this only takes into account a person's purported habits,
not the actual exercise activity that a person engages in each day.
Accordingly, the score is static and does not change in relation to
actual activity performed.
[0006] Such disclosed systems are primarily directed to medical
practitioners for addressing issues in continuity of care and
require input from practitioners in order to produce and maintain
scores. Clearly, while the attention of a medical practitioner is
needed in emergency and critical care situations, cost and resource
factors mean that such systems are usable only in such situations
and such systems do not address the general issues discussed above.
Additionally, the score is only relevant to the particular instant
in time at which it was last updated by the medical
practitioner.
SUMMARY OF THE INVENTION
[0007] According to an aspect of the present invention, there is
provided a computer implemented method for processing private
health related data into a masked numerical score suitable for
publishing. The method comprises receiving data into a memory on a
plurality of intrinsic medical parameters and extrinsic physical
activity parameters of a user. The received data and weighting
factors are stored in the memory. The received data is processed by
executing code in a processor that configures the processor to
apply the weighting factors to the intrinsic medical parameters and
the extrinsic physical activity parameters. The weighting factors
for at least the extrinsic physical activity parameters include a
decay component arranged to reduce the relative weight of the
extrinsic physical activity parameters for a physical activity in
dependence on at least one factor associated with the user. The
processed data concerning the intrinsic medical parameters and the
extrinsic physical activity parameters are transformed by further
code executing in the processor into a masked composite numerical
value in which the code is operative to combine the weighted
parameters in accordance with an algorithm. The masked composite
numerical value is automatically published to a designated group
via a portal (such as a social web site) using code executing in
the processor and free of any human intervention. Meanwhile, the
collected information concerning the intrinsic medical parameters
and the extrinsic physical activity parameters is maintained
private.
[0008] According to a further aspect of such a method as can be
implemented in a particular embodiment thereof, the factor
associated with the user can be an age or an age range of the user
such that the decay component reduces the relative weight of the
extrinsic physical activity parameters for a first user of a first
age or age range differently than a second user of a second age or
age range.
[0009] According to still another aspect of such a method as can be
implemented in a particular embodiment thereof, the published
masked composite numerical value can comprise an average of a group
of users to arrive at a group composite numerical value
determination using further code executing in the processor.
[0010] According to an additional aspect of the present invention,
there is provided a computer implemented health monitoring system
which comprises a communication unit operable to receive data on a
plurality of intrinsic medical parameters and extrinsic physical
activity parameters of a user. A memory is arranged to store the
received data and to store weighting factors. Also, a processor is
arranged to process the received data by executing code that
configures the processor to apply the weighting factors to the
intrinsic medical parameters and the extrinsic physical activity
parameters. The weighting factors for at least the extrinsic
physical activity parameters include a decay component arranged to
reduce the relative weight of the physical activity parameters for
a physical activity in dependence on at least one factor associated
with the user. The processor is further arranged to execute code to
transform the processed data concerning the intrinsic medical
parameters and the extrinsic physical activity parameters into a
masked composite numerical value using the processor by combining
the weighted parameters in accordance with an algorithm. A portal
is arranged to publish the masked composite numerical value to a
designated group while maintaining the collected information
concerning the intrinsic medical parameters and the extrinsic
physical activity parameters private.
[0011] Such a system can preferably be configured so that the
factor associated with the user can be an age or an age range of
the user such that the decay component reduces the relative weight
of the extrinsic physical activity parameters for a first user of a
first age or age range differently than a second user of a second
age or age range.
[0012] An embodiment in accordance with further aspects of the
invention can comprise a system that communicates either the
processed data or the masked composite numerical value to an
exercise machine. The machine works in conjunction with the system
through programming thereat to automatically establish an exercise
program on the basis of the communicated data or the masked
composite numerical value. Preferably, the system so-configured
receives from the exercise machine into its memory activity
information for inclusion among the extrinsic physical activity
parameters.
[0013] Embodiments of the present invention seek to combine data
from multiple medical and non-medical sources in a system and
method that produce a normalized score for a person that takes into
account available medical, physical activity and optionally
lifestyle data (such as diet) in an arrangement that can be
operated and updated in substantially real-time and does not need
frequent access to a medical practitioner. The score and trends
associated with it can be used for various purposes including
triggering alerts as to possible medical issues or repercussions,
providing user feedback, automated motivation and/or goal setting,
training scheduling, automated referrals for medical analysis.
Among the alerts that can be generated are alerts that are
triggered based on monitoring of a composite numerical value of a
health score that is computed, the computed value of which can
cause a feedback communication to be sent to the user (e.g., within
the system portal or by email, SMS, etc.), as a result of code
executing in a processor and without human intervention, if the
monitoring detects a change in the user's score such as due to
decay in value by operation of the algorithm, or reduction in value
due to eating habits, or in fulfillment of goals input into the
system by the user or by a group the user has associated with, or
as part of a non-user-specific goal program that the system can
have to motivate wellness (e.g., good exercise or eating habits).
Embodiments of the present invention apply a weighting factor to
the respective physical activity and/or lifestyle data such that
recent events have a greater impact on the score than those that
occurred further in the past.
[0014] In the described embodiments, a unique health score
computation method is disclosed which masks underlying health
statistics, yet provides a benchmark for a variety of applications.
In one embodiment, a method for collecting and presenting health
related data is provided. The method includes collecting
information concerning a plurality of intrinsic medical parameters
and extrinsic physical activity parameters of a user. The collected
information is stored in a memory and weighting factors are stored
in the memory. The collected information is processed by executing
code in a processor that configures the processor to apply the
weighting factors to the intrinsic medical parameters and extrinsic
physical activity parameters. The collected information concerning
the intrinsic medical parameters and extrinsic physical activity
parameters is transformed into a masked composite numerical value
using the processor by combining the weighted parameters in
accordance with a predetermined algorithm. The masked composite
numerical value is published to a designated group via a portal
while maintaining the collected information concerning the
intrinsic medical parameters and extrinsic physical activity
parameters private.
[0015] Preferred embodiments of the present invention seek to
provide a normalized rating system that can provide an assessment
of the relative health of an individual that can be used as the
basis of a fair comparison to other individuals having different
ages, sex, medical status or lifestyles.
[0016] Various features, aspects and advantages of the invention
can be appreciated from the following Description of Certain
Embodiments of the Invention and the accompanying Drawing
Figures.
DESCRIPTION OF THE DRAWING FIGURES
[0017] FIG. 1 is a schematic block diagram of a local health
information collection and communication system according to a
first implementation of the invention;
[0018] FIG. 1A is a network diagram according to another
implementation of the invention;
[0019] FIG. 2 is a schematic flow diagram according to one
embodiment of the invention;
[0020] FIGS. 3a-3e are screen shots of a user interface according
to one embodiment of the invention;
[0021] FIG. 3f is an illustration of progressions over time of
parameters used to determine the health score in one embodiment of
the invention;
[0022] FIG. 4a is an illustration of a data presentation format
according to one embodiment of the invention;
[0023] FIG. 4b is an illustration of a data presentation format
according to one embodiment of the invention;
[0024] FIG. 4c is an illustration of a data presentation format
according to one embodiment of the invention; and
[0025] FIG. 4d is an illustration of a data presentation format
according to one embodiment of the invention.
DETAILED DESCRIPTION OF CERTAIN EMBODIMENTS OF THE INVENTION
[0026] By way of overview and introduction, the present invention
is described in detail in connection with a distributed system in
which data acquisition, data storage, and data processing are used
to produce a numerical score as a basis for assessing the relative
health of a user.
[0027] In one implementation, a system 100 includes a
computer-based application for the collection of health related
parameters of a user and a user interface 110 for the display of
data. The computer-based application is implemented via a
microcontroller 120 that includes a processor 124, a memory 122 and
code executing therein so as to configure the processor to perform
the functionality described herein. The memory is for storing data
and instructions suitable for controlling the operation of the
processor. An implementation of memory can include, by way of
example and not limitation, a random access memory (RAM), a hard
drive, or a read only memory (ROM). One of the components stored in
the memory is a program. The program includes instructions that
cause the processor to execute steps that implement the methods
described herein. The program can be implemented as a single module
or as a plurality of modules that operate in cooperation with one
another. The program is contemplated as representing a software
component that can be used in connection with an embodiment of the
invention.
[0028] A communication subsystem 125 is provided for communicating
information from the microprocessor 120 to the user interface 110,
such as an external device (e.g., handheld unit or a computer that
is connected over a network to the communication subsystem 125).
Information can be communicated by the communication subsystem 125
in a variety of ways including Bluetooth, WiFi, WiMax, RF
transmission, and so on. A number of different network topologies
can be utilized in a conventional manner, such as wired, optical,
3G, 4G networks, and so on.
[0029] The communication subsystem can be part of a communicative
electronic device including, by way of example, a smart phone or
cellular telephone, a personal digital assistant (PDA), netbook,
laptop computer, and so on. For instance, the communication
subsystem 125 can be directly connected through a device such as a
smartphone such as an iPhone, Google Android Phone, BlackBerry,
Microsoft Windows Mobile enabled phone, and so on, or a device such
as a heart rate or blood pressure monitor (such as those
manufactured by Withings SAS), weight measurement scales (such as
those manufactured by Withings SAS), exercise equipment or the
like. In each instance, the devices each comprise or interface with
a module or unit for communication with the subsystem 125 to allow
information and control signals to flow between the subsystem 125
and the external user interface device 110. In short, the
communication sub-system can cooperate with a conventional
communicative device, or can be part of a device that is dedicated
to the purpose of communicating information processed by the
microcontroller 120.
[0030] When a communicative electronic device such as the types
noted above are used as an external user interface device 110, the
display, processor, and memory of such devices can be used to
process the health related information in order to provide a
numerical assessment. Otherwise, the system 100 can include a
display 140 and a memory 150 that are associated with the external
device and used to support data communication in real-time or
otherwise. More generally, the system 100 includes a user interface
which can be implemented, in part, by software modules executing in
the processor of the microcontroller 120 or under control of the
external device 130. In part, the user interface can also include
an output device such as a display (e.g., the display 140).
[0031] Biosensors 115 can be used to directly collect health
information about a user and report that information. The biosensor
can be placed in contact with the user's body to measure vital
signs or other health related information from the user. For
example, the biosensor can be a pulse meter that is worn by the
user in contact with the user's body so that the pulse of the user
can be sensed, a heart rate monitor, an electrocardiogram device, a
pedometer, a blood glucose monitor or one of many other devices or
systems. The biosensor can include a communication module (e.g.,
communication subsystem 125) so that the biosensor can communicate,
either wired or wirelessly, the sensed data. The biosensor can
communicate the sensed data to the user interface device, which in
turn communicates that information to the microcontroller.
Optionally, the biosensor can directly communicate the sensed the
data to the microprocessor. The use of biosensors provides a degree
of reliability in the data reported because it eliminates user
error associated with manually, self-reported data.
[0032] Alternatively or in addition, the user can self-report his
or her health related information by manually inputting the data.
Thus, in another implementation, as shown in FIG. 1A, health
related data of a person is entered directly into a computer 160
and provided across a network 170 to a server computer 180. (All
computers described herein have at least one processor and a
memory.)
[0033] Regardless of the implementation, the system provides a
means for assigning a numerical value that represents the relative
health of an individual. The numerical value is described herein as
a "health score" and can be used to assess to the individual's
health based on health related information collected from a user.
The health score is calculated based on the collected health
information using an algorithm. The user or the communication
subsystem 125 provides the system the health related information
concerning a number of health parameters. Predetermined weighting
factors are used to assign a relative value of each of the
parameters that are used to calculate the health score. The user's
health score is then calculated by combining the weighted
parameters in accordance with an algorithm. For example, the
parameters can be a person's blood glucose level and body weight. A
weighting factor "a" is applied to the blood glucose data and a
weight factor "b" can be applied to the body weight data. If the
blood glucose data is a more important factor in determining a
person's health than body weight, then the weighting factor "a"
will be larger than weighting factor "b" so that the blood glucose
data has a larger impact on the calculated health score (e.g.,
Healthscore=Glucose*a+(Weight/100)*b). In certain implementations,
the weighting factor is a non-unity value (e.g., greater or less
than one, but not one). Fewer or additional factors can be included
in the calculation of the health score, and an offset value can be
included that is added or subtracted or which modifies the entire
calculation, in certain implementations such as to account for age
or gender as two possible reasons; however, the foregoing is
intended as a non-limiting example of how to calculate a health
score. Other parameters that can be measured and included in the
calculation include blood pressure measurements, height, body mass
index, fat mass, medical conditions such as diabetes, ventricular
hypertrophy, hypertension, irregular heartbeat and fasting glucose
values. Where absent, a parameter can be omitted from the
calculation or it can be estimated from other parameters and/or
values obtained from a sample group of individuals having similar
parameters.
[0034] In addition to intrinsic medical parameters, physical
activity of a user is also taken into account when calculating his
or her health score. Physical activity can be monitored via an
appropriate sensor dependent on the activity. Sensors can include a
GPS unit, an altimeter, a depth meter, a pedometer, a cadence
sensor, a velocity sensor, a heart rate monitor or the like. In the
case of gym-based activities, computerized exercise equipment can
be configured to provide data directly on the program completed by
the user (for example, a so-called elliptical/cross-trainer can
provide far better data on the workout than a user's pedometer
etc). Although automated capture of parameters concerning a user's
physical activity is preferred, a user interface for manual
activity entry is also provided. In this regard, an exercise
machine such as a treadmill, elliptical, stationary bike or weight
lifting machine with a rack of weights or bands can be provided
with a communications interface to communicate with the system
described herein to provide extrinsic physical activity parameters
to the system and to receive and further include a processor
configured to process data from the system so as to automatically
adjust an exercise program at the exercise machine to meet a goal,
challenge, or other objective for that user. Lifestyle data such as
diet, smoking, alcohol consumed and the like can also be collected
and used in calculating the health score. In one embodiment, a
barcode or RFID scanner can be used by a user to capture data on
consumed foodstuffs that is then translated at a remote system,
such as the server 180 or a website in communication with the
server 180, into parameters such as daily calorie, fat and salt
intake. In part, the system relies on such data being provided by
the user while other data can be obtained through data network
connections once permissions and connectivity rights are in
place.
[0035] Physical activity and lifestyle data is tracked over time
and a decay algorithm is applied when calculating its effect on the
health score, as is discussed in more detail below. As such,
physical activity far in the past has a reduced positive effect on
the health score. Preferably, the weighting factors used in the
algorithm for the computation of the health score are adjusted over
time in accordance with a decay component which is arranged to
reduce the relative weight of the parameters that are used in the
calculation. The decay component can itself comprise a weighting
value, but can also comprise an equation that takes into account at
least one factor associated specifically with the user, such as the
user's weight or weight range, age or age range, any medical
conditions known to the system, and any of the other parameters
that may be known to the system, or a curve that is configured in
view of these factors so that a value can be read from the curve as
a function of the values along the axes for that user. In this way,
the decay component can reduce the relative weight of the
parameters used in the health score calculation for a first user
differently than for another user, such as when the first user has
a first age or age range and the second user has a second age or
age range.
[0036] A central system, preferably a database and website that can
be hosted, for example, by the server 180, maintains data on each
user and his or her health score and associated parameters and
their trends over time. The data can be maintained in such a way
that sensitive data is stored independent of human identities, as
understood in the art.
[0037] The calculated health score for each user is then processed
in dependence on a system, group or user profile at the central
system. Depending on the profile settings, the health score and
trends associated can cause various automated actions. For example,
it can cause: triggering of an automated alert; providing user
feedback such as a daily email update; triggering the communication
of automated motivation, warnings and/or goal setting selected to
alleviate a perceived issue; adjustment of a training programme; or
automated referral for medical analysis.
[0038] The user's health score is also provided to a designated
group of recipients via a communication portal. The group of
recipients can comprise selected, other, users of the system (e.g.,
friends and family) so that the health scores of the selected,
other users can be compared against the health score of still
others. In alternative arrangements, all users can see other user's
scores, or the group of recipients can be defined as a specific
health insurance provider so that price quotes can be provided to
insure the individual. Other possibilities are within the scope the
invention.
[0039] Referring now to FIG. 2, a schematic flow diagram according
to one embodiment of the invention is described in support of an
assessment of a person (e.g., a patient or user) to provide a
health score. At step 210, the user initiates the process for the
collection, processing, and publishing of health related data. For
example, a person using a mobile electronic device (e.g. a smart
phone or portable computing device) selects the software
application, which starts the program running on the device
processor, or the user can access an Internet based web page in
which code is executed on a remote processor and served to the
user's local device. An identification module prompts the user to
identify himself and authenticate his identity. This can be
accomplished by prompting the user to enter a user name and
password, or by other means, such as a fingerprint reader, keyfob,
encryption or other mechanism to ensure that identity of the user.
Alternatively, if the user is accessing the system via a personal
electronic device, identification data can be stored in the local
device memory and automatically accessed in order to automatically
confirm the identity of the user.
[0040] At step 220, a data collection module executing on the
processor can prompt the user to provide health related data
corresponding to a number of parameters. In one implementation, one
or more the parameters are provided automatically by the
communication subsystem 125. The parameters can include the user's
body weight, height, age and fitness activity information. Such
measurable medical parameters are intrinsic parameters of the user.
The user's body weight and height provide information about the
user's current state of health. The fitness activity information
corresponds to the amount of exercise the user engages in. This
information is an example of a physically activity parameter that
is an extrinsic parameter of the user. For example, the user can
enter information about his or her daily fitness activities, such
as the amount of time the user engaged in physical activity and the
type of physical activity. If the user went to the gym and
exercised on a bicycle for thirty minutes, for example, that
information is entered into the system. The user's fitness activity
information provides information about the actions that are being
taken by the user in order to improve his or her fitness.
[0041] A user's body weight, height, age and fitness activity
information are just some of the parameters for which information
can be collected. The system can collect and process a multitude of
other parameters that can be indicative of a user's health. For
example, parameters can include blood glucose levels, blood
pressure, blood chemistry data (e.g., hormone levels, essential
vitamin and mineral levels, etc.), cholesterol levels, immunization
data, pulse, blood oxygen content, information concerning food
consumed (e.g., calorie, fat, fiber, sodium content), body
temperature, which are just some of a few possible, non-limiting
examples of parameters that can be collected. Various other
parameters that are indicative of a person's health that can be
reliably measured could be used to calculate a person's health
score.
[0042] The collected health parameter information is stored in a
memory at step 230. At step 240, a weighting module recalls
weighting factors from the memory. The weighting factors can be
multiplication coefficients that are used to increase or decrease
the relative value of each health parameters. A weighting factor is
assigned to each health parameter as shown in the formulas herein.
The weighting factors are used to control the relative values of
the health parameters. Some health parameters are more important
than others in the calculation of the users health score.
Accordingly, weighting factors are applied to the health parameters
increase or decrease the relative affect each factor has in the
calculation of the user's health score. For example, a user's
current body weight can be more important than the amount of
fitness activity the user engages in. In this example, the body
weight parameter would be weighted more heavily by assigning a
larger weighting factor to this parameter. At step 205, the
weighting module applies the recalled weighting factors to the
collected health parameter values to provide weighted health
parameter values. The weighting factor can be zero in which case a
particular parameter has no impact on the health score. The
weighting factor can be a negative value for use in some
algorithms.
[0043] After the parameters have been weighted, the user's health
score is computed at step 260 via a scoring module operating in the
processor. The scoring module combines the weighted parameters
according to an algorithm. In one implementation, the health score
is the average of the user's body mass index (BMI) health score and
the user's fitness health score minus two times the number of years
a person is younger than 95. The algorithm formula for this example
is reproduced below:
Health Score=((BMI Healthscore+Fitness
Healthscore)/2)-2*(95-Age).
The user's BMI Healthscore is a value between 0 and 1000. The BMI
Healthscore is based on the user's BMI, which is calculated based
on the user's weight and height, and how much the user's BMI
deviates from what is considered a healthy BMI. A chart or formula
can be used to normalize the user's BMI information so that
dissimilar information can be combined. A target BMI value is
selected which is assigned a maximum point value (e.g. 1000). The
more the user's BMI deviates from the target value the fewer points
are awarded. The user's Fitness Healthscore is based on the
physical activity or exercise of a person. In one embodiment, it is
the sum of the number of fitness hours (i.e., the amount of time
the user engaged in fitness activities) in the past 365 days where
each hour is linearly aged over that time so that less recent
activity is valued less. The resulting sum is multiplied by two and
is capped at 1000. This normalized the fitness information so that
it can be combined to arrive at the health score. A target daily
average of fitness activity is selected and is awarded the maximum
amount of points (e.g. 1000). The user is awarded fewer points
based on how much less exercise that engage in compared to the
target.
[0044] In another implementation, the health score is determined
from a number of sub-scores that are maintained in parallel beyond
the BMI health score and the fitness health score. Likewise, the
health score can be determined using similar information in a
combinative algorithm as discussed above using different or no age
adjustments.
[0045] Intrinsic medical parameters are processed to determine a
base health score. Extrinsic parameters such as those from physical
exercise are processed to determine a value that is allocated to a
health pool and a bonus pool. The value, preferably expressed in
MET hours, associated with a physical activity is added to both the
health pool and the bonus pool. A daily decay factor is applied to
the bonus pool. Any excess decay that cannot be accommodated by the
bonus pool is then deducted from the health pool. The amount of
decay is determined dependent on the size of the health pool and
bonus pool such that a greater effort is required to maintain a
high health and bonus pool. The health pool value is processed in
combination with the score from the intrinsic medical parameters in
order to calculate the overall health score value. This can be on a
similar basis to the earlier described implementation or it can
include different parameters and weighting factors. In one
embodiment, the health pool value is a logarithm or other
statistical function is applied to age the respective values over
time such that only the most recent activity is counted as being
fully effective to the health/bonus pool. An example user interface
showing the health score, the health reservoir and selected other
measured parameters (as it will be appreciated that many simply
combine to make up the scores) is shown in FIGS. 3a and 3b. Various
sub-scores and their trends are recorded, as is shown in FIG.
3c.
[0046] As will be appreciated, MET hours are kcal expended divided
by kilograms of body weight, i.e. 100 kcal expended by a person of
50 kg is 2 MET h. This is "normalized energy", making the system
fair for persons of all weights. With this method, pools can be the
same size for each per person as energy is normalized for the
person based on his or her body weight.
[0047] In one implementation, each person is assigned a health pool
having a capacity of 300 MET h and a bonus pool having a capacity
of 60 MET h.
[0048] When someone performs activity A, the pools are updated as
follows:
H=min(H+A*alpha,300)
B=min(B+A*(1-alpha),60)
[0049] Where H is the health pool score, B is the bonus pool score,
A is the MET h value for the activity and alpha is a system wide
contestant (selected between 0 and 1) that determines the
proportion in which the activity contributes to the respective
pools.
[0050] The activity is split between the health pool and the bonus
pool. Any excess MET h activity going over the cap of any pool is
discarded. A daily decay value D is applied to the pools as
follows:
D=f(H,B)
B=B-D
If B<0:
D=D+B
B=0
If D<0:
D=0
[0051] The decay is fully applied to the bonus pool, and if the
bonus pool is empty, the remainder is applied to the health pool.
In this embodiment, no pool ever goes below zero.
[0052] The system finds its equilibrium where A equals f(H, B),
i.e. where the average daily activity matches the average daily
decay. The function f(H, B) is highly non-linear with regard to H
and B. In essence, it takes sub-linearly less effort to maintain a
small pool, and super-linearly more effort to maintain a large
pool. This is to make sure that the average person can maintain a,
say, half-full health pool (150, corresponding to a score of 500),
whereas it takes a massively higher effort (typically only
delivered by a professional endurance athlete) to maintain a full
health pool (300, corresponding to a score of 1000). FIG. 3f shows
a simulation of the buffer pool and health reservoir score over
time assuming activity varying between 11.5 and 16 MET h per day
and 2 days off per week. A perfect health reservoir score of 1000
would require 30 MET h activity per day, as can be seen from the
curve in the top right corner of FIG. 3f
[0053] Preferably, the health score is based on a weighted
combination of health factor(s) and the exercise record of the
person over time. The health factors can be updated regularly by
the user. For example, the user can provide health related
information after every event that is tracked and processed by the
system. The user can update after a meal, after exercising, after
weighing himself, etc. In the case of recordal of an activity/event
by a sensor, portable device or the like, the captured/calculated
parameters can be automatically uploaded and used to produce a
revised health score. For example, feedback could be provided
showing the effect of exercise while a user is running, working out
on exercise equipment etc. In selected embodiments, feedback can be
provided to an administrator such as a gym staff member where it is
determined that a user is exceeding a predetermined threshold
(which due to knowledge of their health can be varied respective to
their health score or other recorded data). Accordingly, the health
related data can be updated in a near real-time manner.
[0054] The user can also update the information twice daily, once
daily, or at other periodic times. Moreover, the health score can
be based on an average of the information over time. Fitness
activity, for example, can be averaged over a period of time (e.g.
over a week, month, or year). Averaging data over time will reduce
the impact to the health score caused by fluctuations in data.
Periods in which the data was uncharacteristically high (e.g., the
person was engaging large amount of fitness activity over a short
period of time) or uncharacteristically low (e.g., person engaged
in no fitness activity for a week due to an illness) does not
dramatically affect the health score with averaging over time. The
health related information can be stored in the memory or in a
database accessible by the processor.
[0055] The stored data can also be used to predict future health
scores for a user. A prediction module can analyze past data (e.g.,
fitness habits, eating habits, etc.) to extrapolate a predicted
health score based on an assumption that the user will continue to
act in a predicable manner. For example, if the data shows that a
user has exercised one hour every day for the past thirty days, the
prediction module can predict, in accordance with a prediction
algorithm, that the user will continue to exercise one hour for
each of the next three days. Accordingly, the scoring module can
calculate a predicted health score at the end of the next three
days based on the information from the prediction module. It can
also factor the prediction into other actions. For example, the
system can suggest a more exerting physical activity level or
challenge to someone who has a high health score but is predicted
based on past experience to then take a number of days off for
recuperation. Furthermore, the system can provide encouragement to
the user to maintain a course of activity or modify behavior. For
example, the system can send a message to the user indicating that
if the user increased fitness activity by a certain amount of time,
the health score would go up by a certain amount. This would allow
the user set goals to improve health.
[0056] The use of the health score allows for a relative comparison
of a user's health with that of another person's even though each
person can have very different characteristics, which would make a
direct comparison difficult. For example, a first user (User 1) can
have a very different body composition or engage in very different
fitness activities as compared to a second user (User 2), which
makes direct comparison of the relative health of each user
difficult. The use of the health score makes comparison of the two
users possible with relative ease. In one example, User 1 is
slightly overweight, which would tend to lower User 1's health
score. However, User 1 also engages is large amounts of fitness
activities, thereby raising the user's overall health score. In
contrast, User 2 has an ideal body weight, which would contribute
to a high health score, but engages in very little fitness
activity, thereby lowering the health score. User 1 and User 2 are
very different in terms of their health related parameters.
Accordingly, it would be very difficult to assess and compare the
relative health of User 1 and User 2. In accordance with the
invention, information related to certain health parameters is
collected from User 1 and User 2, which is used to calculate an
overall health score. A comparison of User 1's and User 2's health
score allows for an easy assessment and comparison of the health of
these two users even though they are very different and have very
different habits. Therefore, the health score has significant value
so that members of a group can compare their relative health and so
that other entities (e.g., employers, health care insurers) can
assess the health of an individual. Examples are shown in FIGS. 3d
and 3e in which tabular (current) and graphical (historic, current
and predicted) scores of different users are shown. As can be seen
in FIG. 3e, Katrin is expected to surpass the user (Andre) shortly
unless he improves his lifestyle and performance. In FIG. 3d, the
impact of the decay algorithm is illustrated to show the effect on
the health score of a given user (Andre) and the people he has
identified as friends. As noted, user Andre has a current health
score of 669 which situates this user between friends Irene (health
score 670) and Helle (health score 668). The decay algorithm has
acted on all of the health scores shown in the screen shot of FIG.
3d, as indicated in the ".DELTA. 1 Day" column. More particularly,
most of the friends of Andre have had their health score reduced by
1 point due to the reason of "no activity." A lack of data input to
the system is a basis for the processor executing the decay
algorithm to determine a "no activity" status for a given user. The
one day effect of this status according to the illustrated decay
algorithm for most of the users is a reduction of 1 point in one
day, and a reduction of 5 points over the course of a week. As
such, the decay algorithm can have a tapering, non-linear impact on
an overall health score.
[0057] As illustrated, user Andre has had moderate activity
registered into a memory that is accessible to the system. As a
result, the moderate activity is processed and results in a one day
change (delta) that is positive, and a change that counteracts the
influence of the decay algorithm. Consequently, Andre will be able
to observe, as well as the friends that have access to his
published health score, that he increased his score from 667 to 669
in one day, and from 662 to its present value over the past seven
days as a result of "moderate activity." Moreover, a prediction is
computed using the underlying algorithm and an extrapolation of
data based on most recent reasons (that is, received data) to
increase another 5 points. On the other hand, due to low activity,
but a good diet, Helle in the same time period went down 1 point in
the last day and a total of 1 point in the last 7 days and is
predicted to lose another point if this rate continues. As such,
Helle is provided with feedback by execution of the algorithm and
the outputs provided by the system which can encourage more
activity. On the other hand, Irene has no activity and a poor diet
which results in a more aggressive change to her current health
score and the longer-view historical and predicted impact on her
score. Again, this feedback, which can be provided to users and
their friends or to members of a group of users that have
associated together for a challenge, etc. to provide individual or
team motivation to engage in fitness activities, eat well, and so
on.
[0058] Moreover, the health score provides an indication of the
relative health of the individual without revealing the underlying
data used to calculate the health score, which can be sensitive
information. For example, a user may be uncomfortable revealing his
or her weight, age, or amount of time they spend exercising to
others persons or entities. Persons can be embarrassed to share his
or her weight or the fact that they virtually never go to the gym.
However, since the health score is derived from several factors,
the underlying data used to calculate the score is kept private.
This feature will facilitate the sharing of the user's overall
health because users will not have to disclose private data about
themselves. For example, a person may be slightly overweight, but
she goes to the gym often. Accordingly, that person can receive a
relatively good health score. While the person may not want to
disclose her weight, she can still disclose her health score which
conveys information about her relative health without disclosing
the underlying details. The intrinsic medical parameters (e.g.
weight, height, etc.) and the extrinsic physical activity
parameters (e.g. exercise duration, frequency, intensity, etc.) are
transformed into a masked composite numerical value. The masked
numerical value is published while the collected information
concerning the intrinsic medical parameters and extrinsic physical
activity parameters is maintained private. The underlying intrinsic
medical parameters and extrinsic physical activity parameters are
protected so that a third party is not able to determine those
parameters based on the health score number. This is because the
parameters can vary in many different ways and yet the health score
number could be the same (e.g., a heavier person that exercises
frequently can have the same health score as a person that is not
overweight but does not exercise as frequently). Thus, having the
health score alone does not reveal a person's health related
parameters. Accordingly, the underlying health statistics are
masked, yet the health score can be used as a benchmark to indicate
a person's health for a variety of applications.
[0059] After the scoring module calculates the health score of the
user, at step 270, a publication module recalls from the memory the
designated group of recipients that are authorized to receive the
health score. The group of recipients can be friends or family of
the user, sports teammates, employers, insurers, etc. At step 280,
the publication module causes the health score to be published to
the designated group. In the case that the information is to be
published to a group of friends, the information can be published
to a social networking internet based portal in which access to the
data is limited to those designated members of the group.
[0060] The health parameter data and health scores can be stored
over time, in a memory or other database, so that a user can track
his or her progress. Charts can be generated in order for a user to
track progress and analyze where there can be improvement in
behavior. Moreover, trends can be identified that can lead to the
diagnosis of medical problems and/or eating habits. For example, if
a person's weight is continuing to increase despite the same or
increased amount of fitness activity, the system can trigger or
suggest that they seek certain medical tests (e.g. a thyroid test,
pregnancy test) to determine the cause of the weight gain.
[0061] In certain implementations, the majority of the system is
hosted remotely from the user and the user accesses the system via
a local user interface device. For example the system can be
internet based and the user interacts with a local user interface
device (e.g., personal computer or mobile electronic device) that
is connected to the internet (e.g., via a wire/wireless
communication network) in order to communicate data with the
internet based system. The user uses the local interface device to
access the internet based system in which the memory and software
modules are operating remotely and communicating over the internet
with the local device. The local device is used to communicate data
to the remote processor and memory, in which the data is remotely
stored, processed, transformed into a health score, and then
provided to the designated groups via a restricted access internet
portal. Alternatively, the system can be primarily implemented via
a local device in which the data is locally stored, processed, and
transformed into a health score, which is then communicated to a
data sharing portal for remote publication to the designated
groups.
[0062] The system can be implemented in the form of a social
networking framework that is executed by software modules stored in
memory and operating on processors. The system can be implemented
as a separate, stand alone "health themed" social networking system
or as an application that is integrated with an already existing
social networking system (e.g., Facebook, MySpace, etc.). The user
is provided with a homepage in which the user can enter
information, manage which information is published to designated
groups, and manage the membership of the designated groups. The
homepage includes prompts to the user to enter the health related
information for the each of the various parameters. The user can
enter his or her weight, date of birth, height, fitness activity,
and other health related information. The user's health score is
then calculated. The health score is shared with other users that
are designated as part of a group permitted to have access to that
information. Moreover, the user can view the health score
information of others in the group. Accordingly, the user is able
to compare his or her overall health with the health of others in
the group. Comparison of health scores with others in the group can
provide motivation to the individuals in the group to compete to
improve their health scores. Other information, such as health
tips, medical news, drug information, local fitness events, health
services, advertising and discounts for medical and/or fitness
related supplies and service, issuance of fitness challenges or
health related goals, for example, can be provided via the
homepage.
[0063] In further implementations, the health score can be a
composite of a Metric Health Model score and a Quality of Life
Model score. Combining scores from multiple models provides a more
holistic assessment of a user's health. The Metric Health Model
score assesses a user's health based on relatively easily
quantifiable parameters (e.g., age, sex, weight, etc.) and compares
those numbers to acceptable populations study models. The Quality
of Life Model score focus on a user's self-assessed quality of life
measure based on responses to a questionnaire (i.e., the system
takes into account the user's own assessment of their health and
life quality) because there are correlations between how an
individual "feels" about his or her life and a realistic measure of
health. A combination of the scores from these two models, which
will be discussed in more detail below, provides a more inclusive
and holistic assessment of health.
[0064] The Metric Health Model score is based on medical parameter
information of a user, such as their medical history information,
attributes, physiological metrics, and lifestyle information to the
system. For example, the system can provide the user a
questionnaire to prompt responses (yes/no, multiple choice,
numerical input, etc.) or provide the user with form fields to
complete. Medical history information can include the user's
history of medical conditions and/or the prevalence of medical
conditions in the user's family Examples of medical history
information can include information such as whether the user has
diabetes, has direct family members with diabetes, whether the user
or family members have a history of heart attack, angina, stroke,
or Transient Ischemic Attack, a history of atrial fibrillation or
irregular heartbeat, whether the user or family members have high
blood pressure requiring treatment, whether the user or family
members have hypothyroidism, rheumatoid arthritis, chronic kidney
disease, liver failure, left ventricular hypertrophy, congestive
heart failure, regular use of steroid tablets, etc.
[0065] The Metric Health Model score can also be based on user
attributes. The attributes can include age, sex, ethnicity, height,
weight, waist size, etc. In addition, Metric Health Model score can
be based on physiological metrics of the user. Examples of
physiological metrics can include systolic blood pressure, total
serum cholesterol, high-density lipoprotein (HDL), low-density
lipoprotein (LDL), triglycerides, high-sensitivity C-reactive
protein, fasting blood glucose, etc. The inputs can also include
parameters of a user's lifestyle. For example, lifestyle parameters
can include inputs about whether the user is a smoker (ever smoked,
currently smokes, level of smoking, etc.), how much exercise the
user performs (frequency, intensity, type, etc.), type of diet
(vegetarian, high-protein diet, low-fat diet, high-fiber diet,
fast-food, restaurant, home cooking, processed and pre-packaged
foods, size of meals, frequency of meals, etc.). These are some of
the examples of parameters that can be used to compare the user's
health indicators to survival probability models in order to
calculate the user's Metric Health Model score.
[0066] Survival probability prediction models can be used to
predict the probability that an individual will suffer one or more
serious health events over a given time period. Mathematical models
can estimate this probability from observed population
characteristics. Using observational data on a set of unambiguous
severe health events, such as stroke or cardiac infarction, models
can generate the probability that an individual will suffer one
such event over a given time horizon from a set of measurements of
markers, or predictors, for the event (e.g., information about a
user's medical history, attributes, physiological metrics,
lifestyle, etc. as described above). The time distance between the
moment the predictors are measured, and the target event that is
generated by such models, is referred to as a survival probability,
although it must be understood that not all target events
considered are necessarily fatal.
[0067] These survival probability models are typically derived from
the study of generally large populations that are followed for a
considerable length of time, usually more than ten years, and the
statistics collected on the observation of the target event(s) are
summarized and generalized using mathematical methods. There are a
number of such models that exist that have been extensively
validated and maintained and improved by periodically updating the
model's parameters using new data. Examples of existing models can
include a subset of models developed and maintained by the
Framingham Heart Study (an extensive bibliography on results
obtained from the Framingham Heart study is available at
www.framinghamheartstudy.org/biblio), a subset of the models
developed and maintained by the University of Nottingham and the
QResearch Organization (see, for example, J Hippisley-Cox et al,
Predicting cardiovascular risk in England and Wales: prospective
derivation and validation of QRISK2, BMJ 336: 1475 doi:
10.1136/bmj.39609.449676.25 (Published 23 Jun. 2008)), the ASSIGN
model developed by the University of Dundee (see, for example, H
Tunstall-Pedoe et al, Comparison of the prediction by 27 different
factors of coronary heart disease and death in men and women of the
Scottish heart health study: cohort study; BMJ 1998; 316:1881), the
Reynolds model (see, for example, P M Ridker et al, C-Reactive
Protein and Parental History Improve Global Cardiovascular Risk
Prediction: The Reynolds Risk Score for Men, Circulation 2008; 118;
2243-2251, and Development and Validation of Improved Algorithms
for the Assessment of Global Cardiovascular Risk in Women, JAMA,
Feb. 14, 2007--Vol 297, No. 6), the PROCAM model from the Munster
Heart Study (see, for example, Simple Scoring Scheme for
Calculating the Risk of Acute Coronary Events Based on the 10-Year
Follow-Up of the Prospective Cardiovascular Munster (PROCAM) Study,
Circulation. 2002; 105:310-315), and the SCORE model (see, for
example, R M Conroy et al, Estimation of ten-year risk of fatal
cardiovascular disease in Europe: the SCORE project, European Heart
Journal (2003) 24, 987-1003). Other constituent risk models can
also be included. In addition, precursor models can also be used.
Precursor models predict the development of a first condition (e.g.
high blood pressure), wherein the development of the first
condition is predictive of developing a second condition (e.g.,
heart disease). There are models that generate estimates of the
probability of developing diabetes or high blood pressure, for
example, which are two important predictors of mortality. A high
probability of developing diabetes in five years, for instance,
will independently increase the probability of a serious
cardiovascular event within the next ten years. Several such
precursor models can be included and the inclusion of precursor
models leads to more accurate metric risk models, but more
importantly, also leads to the possible reduction of the risk of
mortality through well-defined modifiable aspects of lifestyle.
[0068] Traditional survival probability models have certain
inherent limitations that result from the procedures used to build
them. In deriving such models, researchers compromise between
accuracy and usability. It is difficult for an inductive model,
meaning a model derived directly from data, to include all possible
predictors. This is in part because not all relevant predictors of
a particular event are known, but also in part because some known
predictors may be difficult or expensive to measure. In fact,
several well-known markers of risk, such as genetic factors, are
often not included in such models. Therefore, several potential and
known predictive metrics can be excluded as covariates when
deriving a given survival model.
[0069] Survival probability models are built using data collected
from a given population, and thus summarize and generalize
morbidity and mortality characteristics of the studied population.
However, such a model might be at variance when compared with risk
estimates derived from other populations. When a given model is
used in a population that differs from the one where the model was
built it often underestimates or overestimates a particular risk
because only a few predictors are often considered, and because
other relevant predictors that may not be included in the model
might very well differ between two populations.
[0070] Given the above discussion, together with basic
probabilistic logic, a judicious combination of models derived for
several different populations will generate a better view of the
risks that an individual picked at random is exposed to, and will
thus be more robust in estimating risks for the population at
large. Furthermore, based on mathematical grounds, under very
general assumptions, certain model combination methods, referred to
as predictor boosting, can improve the accuracy of the constituent
models. In fact, boosting a set of models, when done correctly,
will produce a model with accuracy that is, at worst, equal to that
of the most accurate model in the boosted set.
[0071] Accordingly, the Metric Health Model score can be calculated
by comparing the user's medical parameter information to the
survival probability models. A score, preferably in the range of 0
to 1000, with the top end signifying perfect health and the low
side signifying poor health, can be derived following a two-step
process. First, an overall survival probability is obtained from a
combination of the survival probabilities generated by individual
survival probability models, as described above. Second, the
resulting survival probability, which is a number in the 0 to 1
range, is transformed using a parametric nonlinear mapping function
into the 0 to 1000 range. The parametric mapping function is tuned
so that it is linear, with a high slope, in the region of typical
survival probabilities, and asymptotically slopes off in the low
and high ends of the survival probability distribution. The mapping
function is designed to be strongly reactive to changes in the
typical survival probability region.
[0072] As discussed above, the health score can be composed of the
Metric Health Model score, and also the Quality of Life Model
score. The Quality of Life Model score is based on a user's answers
to a set of questionnaires. The system can include several
different questionnaires with some questions in common. The type of
questionnaires and the type of questions therein presented to the
user can be tailored based on a user's health parameters (i.e.,
user age, other data in the user's medical history, etc.). A
specific questionnaire can be generated and presented to the user
on the basis of information on the user that is known to the
system. The questions can be presented with an appropriate multiple
choice response that the user can check/tick on a form, with no
free-form text is entered by the user to permit easier assessment
of the responses. Other types of responses are possible (e.g.,
rating how true a statement is to the user 1-10). The following
list provides several sample questions (in no particular order) on
a number of health-related quality of life topics that can be used
in a system questionnaire.
[0073] Sample Questions: [0074] How do you rate your quality of
life? [0075] How do you rate your overall health? [0076] How much
do you enjoy life? [0077] To what extent do you feel your life to
be meaningful? [0078] How well are you able to concentrate? [0079]
How safe do you feel in your daily life? [0080] How healthy is your
physical environment? [0081] Are you satisfied with your
appearance? [0082] To what extent do you have the opportunity for
leisure activities? [0083] How much do you need any medical
treatment to function in your daily life? [0084] For how long have
your activities been limited because of your major impairment or
health problem? [0085] Do you need help in handling your personal
care needs because of health problems? [0086] Do you need help in
handling your routine needs because of health problems? [0087] Are
you limited in any way in any activities because of any major
impairment or health problem? [0088] How true or false is each of
the following statements for you?: [0089] I seem to get sick a
little easier than other people [0090] I am as healthy as anybody I
know [0091] I expect my health to get worse [0092] My health is
excellent [0093] Do you suffer from any of the following major
impairment or health problem that limits your activities?: [0094]
Arthritis or rheumatism [0095] Back or neck problem [0096] Cancer
[0097] Depression, anxiety or any emotional problem [0098] Vision
problem [0099] Fractures, bone/joint injury [0100] Hearing problem
[0101] Breathing problem [0102] Walking problem [0103] Other
impairment or problem [0104] During the past 30 days, for about how
many days: [0105] was your physical health not good? [0106] did
pain make it hard for you to do your usual activities, such as
self-care, work, or recreation? [0107] have you felt sad, blue or
depressed? [0108] have you felt worried, tense or anxious? [0109]
have you felt you did not get enough rest or sleep? [0110] have you
felt very healthy and full of energy? [0111] have you been a very
nervous person? [0112] have you felt so down in the dumps that
nothing could cheer you up? [0113] have you felt calm and peaceful?
[0114] did you have a lot of energy? [0115] have you felt
downhearted and blue? [0116] did you feel worn out? [0117] have you
been a happy person? [0118] did you feel tired? [0119] How
satisfied are you with: [0120] your sleep? [0121] your ability to
perform your daily living activities? [0122] your capacity for
work? [0123] yourself? [0124] your personal relationships? [0125]
your sex life? [0126] the support you get from your friends? [0127]
the conditions of your living place? [0128] your access to health
services? [0129] your transport? [0130] Are you limited in any of
the following activities because of your health?: [0131] Vigorous
activities, such as running, lifting heavy objects, participating
in strenuous sports [0132] Moderate activities, such as moving a
table, pushing a vacuum cleaner, bowling, or playing golf [0133]
Lifting or carrying groceries [0134] Climbing several flights of
stairs [0135] Climbing one flight of stairs [0136] Bending,
kneeling or stooping [0137] Walking more than a mile [0138] Walking
several blocks [0139] Walking one block [0140] Bathing or dressing
yourself
[0141] This list above is just a sample of questions that can be
presented to a user. The user's responses to the questions are
assigned a value. For example, each of the multiple choice
responses can be assigned a particular value, and all of the user's
response can be tallied to generate a score. In addition, different
questions and different responses can be weighted differently since
some questions, or the severity of the response, can have a greater
predictor of the user's health. The system can also assign a value
based on the user's response to a combination of questions, because
certain combinations can be more predictive of health. Accordingly,
by assessing the user's responses to the questionnaire a Quality of
Life Model score can be derived. Preferably, the Quality of Life
Model score is a numerical value in the range of 0 to 1000.
[0142] The health score is computed as a weighted average of the
Metric Heath Model score and the Quality of Life Model score. The
health score can presented to the user. The health score can be
presented as a numerical value, as a graphic value (i.e. as a
meter, bar, or slider), or a combination of the both, for example.
Referring to FIG. 3A, the health score is presented by a
combination of a numerical score 302 and a slider 304. The slider
can also be color-coded to indicate the score. The position of the
slider bar 306 indicates the user's score.
[0143] One advantage of the presentation of the health score is
that it is not necessary to present the survival probabilities and
raw metrics to the user. Instead, users are presented with a
standardized score. Preferably, this is true of the overall Metric
Heath Model and Quality of Life scores, but it is also true of the
relevant model inputs. This is done mainly to standardize all
output, in the sense that users do not need to know whether high
values of a particular input variable are good or bad; in all
cases, high scores of any input value lead to higher overall health
score values, and low input variable scores lead to lower overall
values of the health score.
[0144] Furthermore, another advantage of the standardized health
scores is that users can compare health scores against other users.
This allows for comparative bench marking (against friends,
co-workers, etc.) with other users. Such score comparisons can be a
part of a game component of the system in which the user competes
against other users, as will be described in more detail below.
Gaming aspects of the system can be used motivate the user of the
health score system, such as a comparison of scores among
user-selected groups, comparison of individual scores within
configurable subpopulation distributions, time-tracking of scores,
and setting of goals, among others. Referring to FIG. 3B, the users
numerical score 302 and graphical score 306 are presented in
combination with a range of scores 308 from a group (e.g. the
world) so that the user can see how his/her score compares to
others in the group. The gaming incentives can be extended by users
to allow the comparison of health scores between users that can
differ substantially in one or more of several specific input
parameters, such as age, weight, and prior risk conditions. The
system highlights improvements in modifiable user metrics,
particularly in lifestyle components, and these improvements in
score provide user incentives. This allows fair competition between
users of a father and his children, for example, via the health
score. In one aspect, the health score provides equalization
between users of different characteristics and is thus similar to
that of a handicap in some sports. Referring to FIG. 3C, the user's
score 306 is compared to the scores 310a-e of a user selected group
of friends. Referring to FIG. 3D, the user's individual medical
parameters (e.g., the medical data provided as a part the Metric
Health Model) can be compared against other users graphically
without revealing the underlying actual values. The high-density
lipoprotein (HDL) level, low-density lipoprotein (LDL) level,
systolic blood pressure (sBP), diastolic blood pressure (dBP), body
mass index (BMI), and fasting blood glucose (fBG) level are shown
on a graph 312. The user's scores are represented by a line 314,
the user's friends scores are each represented by a different dot
316, and a distribution block 318 for a larger population group
(e.g., Switzerland) is also shown. Thus, the user can compare their
individual parameters to a group of friends and the average for a
large population group.
[0145] Users can input data into the system at the time of an event
(i.e., exercise event, food consumption, blood pressure
measurement, etc.), and see the resulting update of their health
score in real-time. The system can include drill-down capabilities,
allowing users to see the various health score component scores,
including tracking over time and the corresponding trends in all
scores; it also includes the setting of goals on the various
scores.
[0146] As an example of use of the system, upon registration with
the system (e.g., the initial use of the system), a user is
prompted to provide medical history data. The user is also prompted
to respond to a complete Quality of Life questionnaire selected by
the system for the given user based on the medical history and user
parameters supplied by the user. After the registration, at
periodic intervals, users are presented with short subsets (3 to 5
questions) of their custom Quality of Life questionnaire to keep
their responses up to date and track changes. Users can enter
inputs for Metric Health Model at any time, and the system prompts
the user for values that have not been updated for some time.
Inputs to the Metric Health Model can be acquired automatically by
the system by accessing a series of digital measuring devices that
have been integrated into the system (e.g., the system can comprise
a mobile electronic communication device, for example, a smart
phone, that is in wireless communication with a measurement device,
such as a blood glucose monitor, so that parameters can be
measured, transmitted, and stored by the system). These can include
weight, blood glucose, physical activity, and other parameters.
Several or multifunction digital measurement devices can be
included in the system. In the case of medical parameters that are
more difficult to obtain with a home measuring device, such as
serum lipid concentration levels, users are only prompted to
provide the relevant data once per (system) configured time period
(e.g., annually and coinciding with a user's routine physical
medical examination).
[0147] To avoid false scores, the system can include several
algorithms to assess the validity of user inputs. The validation
methods can range from ones based on outlier detection to ones
based on multidimensional likelihood estimators. When the system
detects a possible bad input value it flags it and prompts the user
to either confirm the value or to enter a new one.
[0148] The system can generate all its scores, even when missing
one or more inputs. It does this by imputing the missing value or
values using a variety of statistical methods that range from ones
based on global population statistics, to methods based on the use
of more complicated statistical models that are built into the
platform. However, whenever inputs include imputed values, the
system clearly flags all affected scores, and periodically alerts
the user to provide the missing data. The system can also allow for
score simulation, in which the user can temporarily adjust his or
her parameters so that a user can see how changing certain
parameters (e.g., losing weight) affects the user's score.
[0149] The system can also provide recommendations to the users to
take certain actions that can improve the user's health score.
These recommendations can be very specific when any input variable
is in its danger zone, and more generic when any input variable is
outside its optimal range.
[0150] As discussed above, the health score can be used as a part
of a game or competition aspect of the system. The game aspect
increases the fun element of the system for the user and increases
the user's affinity to continue to use the system. The game aspect
can come in the form of obtaining higher levels based on
achievements, competing against others (e.g., in a league), and/or
completing challenges. The "level" is an overall indication of
progress. The level can be monotonically increasing and will
increase by gaining activity points. Activity points can be gained
from performing numerous activities, such as time spent performing
fitness activities (e.g., exercising), improving one's health
score, improving one's BMI, taking part in discussions on the
system (e.g., the system can be a web-based social networking
platform and discussions or "classes" can be offered to teach
fitness skills). A user's level can be displayed on a user's
profile and in discussion posts so that other users can see each
other's level. A user's level status can also provide access to
specific items, system features and functionality, or rewards
(e.g., branded apparel).
[0151] Users can also compete within leagues in the system. The
leagues are composed of groups of users and the users within the
league can compete against each other (as part of a team or
individually). The leagues can compete for a limited time (e.g.,
monthly) and the leagues can be designated based on the level of
the users (using the level of the user as discussed above), the
type of activity being performed in the league, and the geographic
region of the users. For example, one particular league can be the
"bronze" (level) "mountain biking" (sport) "Greater Zurich Area"
(region) league and a user's success in this league is measured by
the distance traveled and elevation climbed (measured quantity).
Thus, bronze level users living in the Greater Zurich Area that are
interested in mountain biking can compete in this league. Limiting
the leagues to a particular region gives the users something to
relate with and all the users can share in common, and further
allows users to meet face to face (e.g., for group exercise
events). One issue with one big international league is that such a
league may seem anonymous, crowded and meaningless to some users
(members competing against members residing on completely different
continents with language barriers can inhibit group or team
mentalities). Limiting leagues to particular level brackets
equalizes the playing field for users of particular skill levels.
Quantities to be measured to determine performance in the league
can include distance (horizontal, vertical) and duration of fitness
activity performed, for example. Users can also form teams within
the leagues. Team leagues work in the same way as the leagues
outlined above, however the ranking is based on the team's overall
performance. Teams increase the communal aspect of participation in
the activity. Teams can be fixed in size (e.g., 2, 3, 5, 10, etc.
users).
[0152] Users can also be presented by the system with challenges to
complete. The challenges can set forth a time period for completion
of a goal. The goals of the challenge can be, for example,
healthscore improvement (normalized), completion of sport-related
activity parameters (e.g., total distance, total climbing, etc.),
or completion of a sport-related activity within a specific period
of time (e.g., complete six minute mile on a specific route). The
challenge can be public and any user can participate, or limited to
a group (e.g. friends, co-workers, social group, etc.) As an
example, a particular public challenge can be an inline skating
challenge in New York City for the route around the Central Park
Loop measuring the time taken for completion. Public challenges can
be generated automatically by the system or by system
administrators. Group challenges can be issued by group members.
Challenges provide strong appointment dynamics, encouraging users
to commit to exercise. Challenges (typically) have a lower time
commitment than leagues. Route selection can be automated with the
community. In a first step, the community can publish routes on the
system platform (e.g., a social networking type website); in a
second step, the system selects popular routes (i.e. routes with
high user activity) as weekly challenges. Route validation is done
by GPS tracking. Challenges can be safety screened to prevent the
promotion of unduly risky challenge activities, such mountain
biking dangerous downhill routes.
[0153] The league and challenge systems provide opportunities to
grant achievements. Achievement status indications can be collected
and displayed on a user's profile. Achievements are much like a
trophy, medal, or award provided to the user for completing
challenges and/or succeeding in a league activity. Many different
achievements are possible, such as related to the number of friends
the user has on the system (community participation), achievements
related to the time, intensity, and number of fitness activities
engaged in (level of fitness participation), achievements related
to specific sport activities (e.g., distance run), the frequency a
user measures their parameters (e.g., weight) in order to keep the
system up to date, the amount of weight lost, or the ability to
maintain ones BMI, for example. The following list is an exemplary
set of achievements and the activities required to earn the
achievements:
[0154] Exemplary Achievement List: [0155] Challenger: Take part in
a public challenge. [0156] Accomplished Challenger: Take part in 10
public challenges. [0157] Champion: Win a challenge. [0158]
Multi-sport Champion: Win a public challenge in two different
sports. [0159] International Challenger: Take part in a public
challenge in two different countries. [0160] International
Champion: Win a public challenge in two different countries. [0161]
World Wide Challenger: Take part in a public challenge on each
continent. [0162] World Wide Champion: Win a public challenge on
each continent.
[0163] Other aspects of the challenge and league systems are that
the systems can be tied to marketing opportunities. For example,
marketers can sponsor prizes for the winners of a challenge. The
prize can be related to the challenge (e.g., gift certificate to
health food score for winner of weight loss challenge). In
addition, challenge routes can be selected to direct users to
certain areas to increase tourism or to begin/end at selected
destinations (e.g., bike challenge begins in front of sports
equipment store).
[0164] One advantage of the system is that it provides users and
groups of users with benchmarking capabilities. It allows other
groups, such as insurance carriers or employers, to assess the
relative health of individuals in order to determine the health
related risks of each individual. Accordingly, users can compare
themselves against others in order to assess their comparative
health level amongst a group of friends. Insurance carriers can use
the health score information to set premiums for an individual or a
group of individuals (e.g. employees of a company). In other
implementations, health scores can be provided for a group based on
the health scores of the individuals in the group. For example, a
health score can be calculated for a company based on its employees
so that an insurance carrier can set premiums based on the health
score of the company compared to other companies. In further
applications, the health score can be used for assessing the health
of professional athletes to determine the athlete's real market
value. Vast amounts of money and resources are invested in athletes
at all levels in professional sports. A large component of the
decision about investing in an athlete is based on the past
performance of the athlete. Other factors can include past physical
injury history and the athlete submitting to a physical exam before
the deal is complete. The health score can be used as an indicator
of the athlete's current health and used as a predictor of the
athletes future performance. If the athlete's health score were
low, this can indicate that the athlete is more prone to suffering
an injury or that physical performance will diminish. Accordingly,
the health score can form a basis for a decision on whether to
invest in an athlete. The health scores could also be used as a
predictor of the outcome of a particular game played between two
teams. For example, the health scores of the individual team
members can be aggregated in order to provide a team health score.
A comparison of the team health scores can be indicative of the
likely outcome of the game between the two teams (e.g., the team
with highest health score may be more likely to win). Such
information can be used in gaming contexts such as fantasy sports
teams, or in order to set odds for sports betting. The health score
could be used for club competitions (e.g., group health improvement
competitions, advertising based on a person's health score, gaming,
tv/internet, etc.
[0165] Thus, in a broad aspect, a method according to the invention
can be understood as collecting health related information,
processing the information into a health score, and publishing the
health score is provided. A system for implementing the method can
include a computer having a processor, memory, and code modules
executing in the processor for the collection, processing, and
publishing of the information. Information concerning a plurality
of health related parameters of a user is collected, particularly,
both intrinsic values concerning the measurable, medical parameters
of at least one natural person, and the extrinsic values concerning
the activities of each such person(s) such as the exercise
performed, the type of job the person has and the amount of
physical work associated with the job (e.g. sedentary, desk job
versus active, manual labor intensive job) and/or the calories/food
consumed. Weighting factors are applied to the health related
parameter in order control the relative affect each parameter has
on the user's calculated health score. The health score is computed
using the processor by combining the weighted parameters in
accordance with an algorithm. The health score is published to a
designated group via a portal. In one implementation, the portal is
an internet based information sharing forum.
[0166] As such, the invention can be characterized by the following
points in a method for collecting and presenting health related
data:
[0167] collecting information concerning a plurality of health
related parameters of a user;
[0168] storing the collected information in a memory;
[0169] storing weighting factors in the memory;
[0170] processing the collected information by executing code in a
processor that configures the processor to apply the weighting
factors to the health related parameters;
[0171] computing a health score using the processor by combining
the weighted parameters in accordance with an algorithm; and
[0172] providing the health score to a designated group via a
portal.
[0173] The methods described herein have been described in
connection with flow diagrams that facilitate a description of the
principal processes; however, certain blocks can be invoked in an
arbitrary order, such as when the events drive the program flow
such as in an object-oriented program implementation. Accordingly,
the flow diagrams are to be understood as example flows such that
the blocks can be invoked in a different order than as
illustrated.
[0174] While the invention has been described in connection with
certain embodiments thereof, the invention is not limited to the
described embodiments but rather is more broadly defined by the
recitations in any claims that follow and equivalents thereof.
* * * * *
References